novita
Pass
Audited by Gen Agent Trust Hub on Aug 18, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill initiates the installation of the
novita-cliPython package during setup. This package is the primary tool provided by the vendor, novitalabs, to interact with their API. - [COMMAND_EXECUTION]: The skill frequently executes the
novitacommand with various subcommands to perform AI tasks, manage GPU instances, and handle files. The execution parameters are derived from user requests and predefined templates. - [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests user-controlled strings and embeds them into prompts for AI generation subcommands.
- Ingestion points: User requests for specific workplace visuals, avatars, or UI designs as described in the Decision Guide in
SKILL.md. - Boundary markers: The templates located in
references/easy-image/templates/use curly-brace placeholders (e.g.,{subject},{organization}) but do not employ explicit delimiters or system instructions to prevent the agent from following commands embedded within the user-provided text. - Capability inventory: The agent utilizes the
novitaCLI for image generation, video creation, and audio synthesis, as mapped inreferences/commands.md. - Sanitization: There is no evidence of input sanitization, filtering, or validation performed on user-supplied strings before they are interpolated into the final prompt sent to the model.
Audit Metadata